NDA 20-905/S-006 NDA 20-905/S-007 Page 4 Rev. XXX ARAVA Tablets (leflunomide) 10 mg, 20 mg, 100 mg Rx only

CONTRAINDICATIONS AND WARNINGS PREGNANCY MUST BE EXCLUDED BEFORE THE START OF TREATMENT WITH ARAVA. ARAVA IS CONTRAINDICATED IN PREGNANT WOMEN, OR WOMEN OF CHILDBEARING POTENTIAL WHO ARE NOT USING RELIABLE CONTRACEPTION. (SEE CONTRAINDICATIONS AND WARNINGS.) PREGNANCY MUST BE AVOIDED DURING ARAVA TREATMENT OR PRIOR TO THE COMPLETION OF THE DRUG ELIMINATION PROCEDURE AFTER ARAVA TREATMENT. DESCRIPTION ARAVA (leflunomide) is a pyrimidine synthesis inhibitor. The chemical name for leflunomide is N-(4´-trifluoromethylphenyl)-5-methylisoxazole-4-carboxamide. It has an empirical formula C12H9F3N2O2, a molecular weight of 270.2 and the following structural formula:

ARAVA is available for oral administration as tablets containing 10, 20, or 100 mg of active drug. Combined with leflunomide are the following inactive ingredients: colloidal silicon dioxide, crospovidone, hypromellose, lactose monohydrate, magnesium stearate, polyethylene glycol, povidone, starch, talc, titanium dioxide, and yellow ferric oxide (20 mg tablet only). CLINICAL PHARMACOLOGY Mechanism of Action Leflunomide is an isoxazole immunomodulatory agent which inhibits dihydroorotate dehydrogenase (an enzyme involved in de novo pyrimidine synthesis) and has antiproliferative activity. Several in vivo and in vitro experimental models have demonstrated an anti-inflammatory effect. Pharmacokinetics Following oral administration, leflunomide is metabolized to an active metabolite A77 1726 (hereafter referred to as M1) which is responsible for essentially all of its activity in vivo. Plasma levels of leflunomide are occasionally seen, at very low levels. Studies of the pharmacokinetics of leflunomide have primarily examined the plasma concentrations of this active metabolite.

O N

CH3

CO NH CF3

NDA 20-905/S-006 NDA 20-905/S-007 Page 5

Absorption Following oral administration, peak levels of the active metabolite, M1, occurred between 6 - 12 hours after dosing. Due to the very long half-life of M1 (~2 weeks), a loading dose of 100 mg for 3 days was used in clinical studies to facilitate the rapid attainment of steady-state levels of M1. Without a loading dose, it is estimated that attainment of steady-state plasma concentrations would require nearly two months of dosing. The resulting plasma concentrations following both loading doses and continued clinical dosing indicate that M1 plasma levels are dose proportional.

Table 1. Pharmacokinetic Parameters for M1 after Administration of Leflunomide at Doses of 5, 10, and 25 mg/day for 24 Weeks to Patients (n=54) with Rheumatoid Arthritis (Mean ± SD) (Study YU204) Maintenance (Loading) Dose

Parameter 5 mg (50 mg) 10 mg (100 mg) 25 mg (100 mg) C24 (Day 1) (µg/mL)1 4.0 ± 0.6 8.4 ± 2.1 8.5 ± 2.2 C24 (ss) (µg/mL)2 8.8 ± 2.9 18 ± 9.6 63 ± 36 t1/2(DAYS) 15 ± 3 14 ± 5 18 ± 9 1 Concentration at 24 hours after loading dose 2 Concentration at 24 hours after maintenance doses at steady state

Relative to an oral solution, ARAVA tablets are 80% bioavailable. Co-administration of leflunomide tablets with a high fat meal did not have a significant impact on M1 plasma levels. Distribution M1 has a low volume of distribution (Vss = 0.13 L/kg) and is extensively bound (>99.3%) to albumin in healthy subjects. Protein binding has been shown to be linear at therapeutic concentrations. The free fraction of M1 is slightly higher in patients with rheumatoid arthritis and approximately doubled in patients with chronic renal failure; the mechanism and significance of these increases are unknown. Metabolism Leflunomide is metabolized to one primary (M1) and many minor metabolites. Of these minor metabolites, only 4-trifluoromethylaniline (TFMA) is quantifiable, occurring at low levels in the plasma of some patients. The parent compound is rarely detectable in plasma. At the present time the specific site of leflunomide metabolism is unknown. In vivo and in vitro studies suggest a role for both the GI wall and the liver in drug metabolism. No specific enzyme has been identified as the primary route of metabolism for leflunomide; however, hepatic cytosolic and microsomal cellular fractions have been identified as sites of drug metabolism.

NH

F3CCH3

C C C N

O

OH

A77 1726 (M1)

NDA 20-905/S-006 NDA 20-905/S-007 Page 6 Elimination The active metabolite M1 is eliminated by further metabolism and subsequent renal excretion as well as by direct biliary excretion. In a 28 day study of drug elimination (n=3) using a single dose of radiolabeled compound, approximately 43% of the total radioactivity was eliminated in the urine and 48% was eliminated in the feces. Subsequent analysis of the samples revealed the primary urinary metabolites to be leflunomide glucuronides and an oxanilic acid derivative of M1. The primary fecal metabolite was M1. Of these two routes of elimination, renal elimination is more significant over the first 96 hours after which fecal elimination begins to predominate. In a study involving the intravenous administration of M1, the clearance was estimated to be 31 mL/hr. In small studies using activated charcoal (n=1) or cholestyramine (n=3) to facilitate drug elimination, the in vivo plasma half-life of M1 was reduced from >1 week to approximately 1 day (see PRECAUTIONS - General - Need for Drug Elimination). Similar reductions in plasma half-life were observed for a series of volunteers (n=96) enrolled in pharmacokinetic trials who were given cholestyramine. This suggests that biliary recycling is a major contributor to the long elimination half-life of M1. Studies with both hemodialysis and CAPD (chronic ambulatory peritoneal dialysis) indicate that M1 is not dialyzable. Special Populations Age and Gender. Neither age nor gender has been shown to cause a consistent change in the in vivo pharmacokinetics of M1. Smoking. A population based pharmacokinetic analysis of the phase III data indicates that smokers have a 38% increase in clearance over non-smokers; however, no difference in clinical efficacy was seen between smokers and nonsmokers. Chronic Renal Insufficiency. In single dose studies in patients (n=6) with chronic renal insufficiency requiring either chronic ambulatory peritoneal dialysis (CAPD) or hemodialysis, neither had a significant impact on circulating levels of M1. The free fraction of M1 was almost doubled, but the mechanism of this increase is not known. In light of the fact that the kidney plays a role in drug elimination, and without adequate studies of leflunomide use in subjects with renal insufficiency, caution should be used when ARAVA is administered to these patients. Hepatic Insufficiency. Studies of the effect of hepatic insufficiency on M1 pharmacokinetics have not been done. Given the need to metabolize leflunomide into the active species, the role of the liver in drug elimination/recycling, and the possible risk of increased hepatic toxicity, the use of leflunomide in patients with hepatic insufficiency is not recommended. Drug Interactions In vivo drug interaction studies have demonstrated a lack of a significant drug interaction between leflunomide and tri-phasic oral contraceptives, and cimetidine. In vitro studies of protein binding indicated that warfarin did not affect M1 protein binding. At the same time M1 was shown to cause increases ranging from 13 - 50% in the free fraction of diclofenac, ibuprofen and tolbutamide at concentrations in the clinical range. In vitro studies of drug metabolism indicate that M1 inhibits CYP 450 2C9, which is responsible for the metabolism of phenytoin, tolbutamide, warfarin and many NSAIDs. M1 has been shown to inhibit the formation of 4´-hydroxydiclofenac from diclofenac in vitro. The clinical significance of these findings with regard to phenytoin and tolbutamide is unknown, however, there was extensive concomitant use of NSAIDs in the clinical studies and no differential effect was observed. (see PRECAUTIONS – Drug Interactions). Methotrexate. Coadministration, in 30 patients, of ARAVA (100 mg/day x 2 days followed by 10 - 20 mg/day) with methotrexate (10 - 25 mg/week, with folate) demonstrated no pharmacokinetic interaction between the two drugs. However, co-administration increased risk of hepatotoxicity (see PRECAUTIONS - Drug Interactions–Hepatotoxic Drugs).

NDA 20-905/S-006 NDA 20-905/S-007 Page 7 Rifampin. Following concomitant administration of a single dose of ARAVA to subjects receiving multiple doses of rifampin, M1 peak levels were increased (~40%) over those seen when ARAVA was given alone. Because of the potential for ARAVA levels to continue to increase with multiple dosing, caution should be used if patients are to receive both ARAVA and rifampin. CLINICAL STUDIES The efficacy of ARAVA in the treatment of rheumatoid arthritis (RA) was demonstrated in three controlled trials showing reduction in signs and symptoms, and inhibition of structural damage. In two placebo controlled trials, efficacy was demonstrated for improvement in physical function.

1. Reduction of signs and symptoms Relief of signs and symptoms was assessed using the American College of Rheumatology (ACR)20 Responder Index, a composite of clinical, laboratory, and functional measures in rheumatoid arthritis. An “ACR20 Responder” is a patient who had >20% improvement in both tender and swollen joint counts and in 3 of the following 5 criteria: physician global assessment, patient global assessment, functional ability measure [Modified Health Assessment Questionnaire (MHAQ)], visual analog pain scale, and erythrocyte sedimentation rate or C-reactive protein. An “ACR20 Responder at Endpoint” is a patient who completed the study and was an ACR20 Responder at the completion of the study.

2. Inhibition of structural damage Inhibition of structural damage compared to control was assessed using the Sharp Score (Sharp, JT. Scoring Radiographic Abnormalities in Rheumatoid Arthritis, Radiologic Clinics of North America, 1996; vol. 34, pp. 233-241), a composite score of X-ray erosions and joint space narrowing in hands/wrists and forefeet.

3. Improvement in physical function Improvement in physical function was assessed using the Health Assessment Questionnaire (HAQ) and the Medical Outcomes Survey Short Form (SF-36). In all Arava monotherapy studies, an initial loading dose of 100 mg per day for three days only was used followed by 20 mg per day thereafter. US301 Study US301, a 2 year study, randomized 482 patients with active RA of at least 6 months duration to leflunomide 20 mg/day (n=182), methotrexate 7.5 mg/week increasing to 15 mg/week (n=182), or placebo (n=118). All patients received folate 1 mg BID. Primary analysis was at 52 weeks with blinded treatment to 104 weeks. Overall, 235 of the 508 randomized treated patients (482 in primary data analysis and an additional 26 patients), continued into a second 12 months of double-blind treatment (98 leflunomide, 101 methotrexate, 36 placebo). Leflunomide dose continued at 20 mg/day and the methotrexate dose could be increased to a maximum of 20 mg/week. In total 190 patients (83 leflunomide, 80 methotrexate, 27 placebo) completed 2 years of double-blind treatment.

NDA 20-905/S-006 NDA 20-905/S-007 Page 8 The rate and reason for withdrawal is summarized in table 2.

Table 2: Withdrawals in US301 n(%) patients Leflunomide Placebo Methotrexate 190 128 190 Withdrawals in Year-1 Lack of efficacy 33 (17.4) 70 (54.7) 50 (26.3) Safety 44 (23.2) 12 (9.4) 22 (11.6) Other1 15 (7.9) 10 (7.8) 17 (9.0) Total 92 (48.4) 92 (71.9) 89 (46.8) Patients entering Year 2 98 36 101 Withdrawals in Year-2 Lack of efficacy 4 (4.1) 1 (2.8) 4 (4.0) Safety 8 (8.2) 0 (0.0) 10 (9.9) Other1 3 (3.1) 8 (22.2) 7 (6.9) Total 15 (15.3) 9 (25.0) 21 (20.8)

1 Includes: lost to follow up, protocol violation, noncompliance, voluntary withdrawal, investigator discretion. MN301/303/305 Study MN301 randomized 358 patients with active RA to leflunomide 20 mg/day (n=133), sulfasalazine 2.0 g/day (n=133), or placebo (n=92). Treatment duration was 24 weeks. An extension of the study was an optional 6-month blinded continuation of MN301 without the placebo arm, resulting in a 12-month comparison of leflunomide and sulfasalazine (study MN303). Of the 168 patients who completed 12 months of treatment in MN301 and MN303, 146 patients (87%) entered a 1-year extension study of double blind active treatment (MN305; 60 leflunomide, 60 sulfasalazine, 26 placebo/ sulfasalazine). Patients continued on the same daily dosage of leflunomide or sulfasalazine that they had been taking at the completion of MN301/303. A total of 121 patients (53 leflunomide, 47 sulfasalazine, 21 placebo/sulfasalazine) completed the 2 years of double-blind treatment.

NDA 20-905/S-006 NDA 20-905/S-007 Page 9 Patient withdrawal data in MN301/303/305 is summarized in table 3.

Table 3: Withdrawals in study MN301/303/305

n(%) patients Leflunomide Placebo Sulfasalazine 133 92 133 Withdrawals in MN301 (Mo 0-6) Lack of efficacy 10 (7.5) 29 (31.5) 14 (10.5) Safety 19 (14.3) 6 (6.5) 25 (18.8) Other1 8 (6.0) 6 (6.5) 11 (8.3) Total 37 (27.8) 41 (44.6) 50 (37.6) Patients entering MN303 80 76 Withdrawals in MN303 (Mo 7-12) Lack of efficacy 4 (5.0) 2 (2.6) Safety 2 (2.5) 5 (6.6) Other1 3 (3.8) 1 (1.3) Total 9 (11.3) 8 (10.5) Patients entering MN305 60 60 Withdrawals in MN305 (Mo 13-24) Lack of efficacy 0 (0.0) 3 (5.0) Safety 6 (10.0) 8 (13.3) Other1 1 (1.7) 2 (3.3) Total 7 (11.7) 13 (21.7) 1 Includes: lost to follow up, protocol violation, noncompliance, voluntary withdrawal, investigator discretion.

MN302/304 Study MN302 randomized 999 patients with active RA to leflunomide 20 mg/day (n=501) or methotrexate at 7.5 mg/week increasing to 15 mg/week (n=498). Folate supplementation was used in 10% of patients. Treatment duration was 52 weeks. Of the 736 patients who completed 52 weeks of treatment in study MN302, 612 (83%) entered the double-blind, 1-year extension study MN304 (292 leflunomide, 320 methotrexate). Patients continued on the same daily dosage of leflunomide or methotrexate that they had been taking at the completion of MN302. There were 533 patients (256 leflunomide, 277 methotrexate) who completed 2 years of double-blind treatment.

NDA 20-905/S-006 NDA 20-905/S-007 Page 10 Patient withdrawal data in MN302/304 is summarized in table 4.

Table 4: Withdrawals in MN302/304

n(%) patients Leflunomide Methotrexate 501 498 Withdrawals in MN302 (Year-1) Lack of efficacy 37 (7.4) 15 (3.0) Safety 98 (19.6) 79 (15.9) Other1 17 (3.4) 17 (3.4) Total 152 (30.3) 111 (22.3) Patients entering MN304 292 320 Withdrawals in MN304 (Year-2) Lack of efficacy 13 (4.5) 9 (2.8) Safety 11 (3.8) 22 (6.9) Other1 12 (4.1) 12 (3.8) Total 36 (12.3) 43 (13.4)

1 Includes: lost to follow up, protocol violation, noncompliance, voluntary withdrawal, investigator discretion.

Clinical Trial Data

1. Signs and symptoms Rheumatoid Arthritis The ACR20 Responder at Endpoint rates are shown in Figure 1. ARAVA was statistically significantly superior to placebo in reducing the signs and symptoms of RA by the primary efficacy analysis, ACR20 Responder at Endpoint, in study US301 (at the primary 12 months endpoint) and MN301 (at 6 month endpoint). ACR20 Responder at Endpoint rates with ARAVA treatment were consistent across the 6 and 12 month studies (41 - 49%). No consistent differences were demonstrated between leflunomide and methotrexate or between leflunomide and sulfasalazine. ARAVA treatment effect was evident by 1 month, stabilized by 3 - 6 months, and continued throughout the course of treatment as shown in Figure 2.

NDA 20-905/S-006 NDA 20-905/S-007 Page 11 Figure 1

% ACR 20 Responder at Endpoint

-20

-10

0

10

20

30

40

50

60

70

1

Perc

enta

ge

L L LM MP S P

40.7

35.0

18.6

48.544.7

28.6

43.4

56.6

Trial US301(12 months)

Trial MN301(6months)

Trial MN302(12 months)

L=Leflunomide, M=Methotrexate, P=Placebo, S=Sulfasalazine

Comparisons 95%Confidence Interval p Value

US301 Leflunomide vs. Placebo (12, 32)

NDA 20-905/S-006 NDA 20-905/S-007 Page 12 ACR50 and ACR70 Responders are defined in an analogous manner to the ACR20 Responder, but use improvements of 50% or 70%, respectively (Table 5). Mean change for the individual components of the ACR Responder Index are shown in Table 6.

Table 5. Summary of ACR Response Rates* Study and Treatment Group ACR20 ACR50 ACR70

Placebo-Controlled Studies US301 (12 months)

Leflunomide (n=178)† 52.2‡ 34.3‡ 20.2‡ Placebo (n=118) † 26.3 7.6 4.2

Methotrexate (n=180)† 45.6 22.8 9.4 MN301(6 months)

Leflunomide (n=130)† 54.6‡ 33.1‡ 10.0§ Placebo (n=91)† 28.6 14.3 2.2

Sulfasalazine (n=132)† 56.8 30.3 7.6 Non-Placebo Active-Controlled Studies MN302 (12 months)

Leflunomide (n=495)† 51.1 31.1 9.9 Methotrexate (n=489)† 65.2 43.8 16.4

* Intent to treat (ITT) analysis using last observation carried forward (LOCF) technique for patients who discontinued early. † N is the number of ITT patients for whom adequate data were available to calculate the indicated rates. ‡ p

NDA 20-905/S-006 NDA 20-905/S-007 Page 13

Table 6. Mean Change in the Components of the ACR Responder Index* Components Placebo-Controlled Studies Non-placebo

Controlled Study US301

(12 months) MN301 Non-US

(6 months) MN302 Non-US

(12 months) Leflu-

nomide Metho- trexate

Placebo Leflu- nomide

Sulfa- salazine

Placebo Leflu- nomide

Metho- trexate

Tender joint count1 -7.7 -6.6 -3.0 -9.7 -8.1 -4.3 -8.3 -9.7 Swollen joint count1 -5.7 -5.4 -2.9 -7.2 -6.2 -3.4 -6.8 -9.0 Patient global assessment2 -2.1 -1.5 0.1 -2.8 -2.6 -0.9 -2.3 -3.0 Physician global assessment2

-2.8 -2.4 -1.0 -2.7 -2.5 -0.8 -2.3 -3.1

Physical function/disability (MHAQ/HAQ)

-0.29

-0.15

0.07

-0.50

-0.29

-0.04

-0.37

-0.44

Pain intensity2 -2.2 -1.7 -0.5 -2.7 -2.0 -0.9 -2.1 -2.9 Erythrocyte Sedimentation rate

-6.26 -6.48 2.56 -7.48 -16.56 3.44 -10.12 -22.18

C-reactive protein -0.62 -0.50 0.47 -2.26 -1.19 0.16 -1.86 -2.45 Not included in the ACR Responder Index Morning Stiffness (min) -101.4 -88.7 14.7 -93.0 -42.4 -6.8 -63.7 -86.6 * Last Observation Carried Forward; Negative Change Indicates Improvement 1 Based on 28 joint count 2 Visual Analog Scale - 0=Best; 10=Worst

Maintenance of effect After completing 12 months of treatment, patients continuing on study treatment were evaluated for an additional 12 months of double-blind treatment (total treatment period of 2 years) in studies US301, MN305, and MN304. ACR Responder rates at 12 months were maintained over 2 years in most patients continuing a second year of treatment. Improvement from baseline in the individual components of the ACR responder criteria was also sustained in most patients during the second year of Arava treatment in all three trials.

2. Inhibition of structural damage The change from baseline to endpoint in progression of structural disease, as measured by the Sharp X-ray score, is displayed in Figure 3. ARAVA was statistically significantly superior to placebo in inhibiting the progression of disease by the Sharp Score. No consistent differences were demonstrated between leflunomide and methotrexate or between leflunomide and sulfasalazine.

NDA 20-905/S-006 NDA 20-905/S-007 Page 14 Figure 3

L= Leflunomide; M=methotrexate; S=sulfasalazine; P=placebo

Comparisons 95% Confidence Interval p Value US301 Leflunomide vs. Placebo (-4.0, -1.1) 0.0007 Methotrexate vs. Placebo (-2.6, -0.2) 0.0196 Leflunomide vs. Methotrexate (-2.3, 0.0) 0.0499 MN301 Leflunomide vs. Placebo (-6.2, -1.8) 0.0004 Sulfasalazine vs. Placebo (-6.9, -0.2) 0.0484 Leflunomide vs. Sulfasalazine (-3.3, 1.2) NS MN302 Leflunomide vs. Methotrexate (-2.2, 7.4) NS

3. Improvement in physical function The Health Assessment Questionnaire (HAQ) assesses a patient’s physical function and degree of disability. The mean change from baseline in functional ability as measured by the HAQ Disability Index (HAQ DI) in the 6 and 12 month placebo and active controlled trials is shown in Figure 4. ARAVA was statistically significantly superior to placebo in improving physical function. Superiority to placebo was demonstrated consistently across all eight HAQ DI subscales (dressing, arising, eating, walking, hygiene, reach, grip and activities) in both placebo controlled studies. The Medical Outcomes Survey Short Form 36 (SF-36), a generic health-related quality of life questionnaire, further addresses physical function. In US301, at 12 months, ARAVA provided statistically significant improvements compared to placebo in the Physical Component Summary (PCS) Score.

NDA 20-905/S-006 NDA 20-905/S-007 Page 15 Figure 4

Change in Functional Ability Measure*

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

1

Impr

ovem

ent

L L LM MP PS

-0.45

-0.26

-0.03

-0.56

-0.37

-0.08

-0.44

-0.54

Trial US301(12 months)

Trial MN301(6 months)

Trial MN302(12 months)

* as measured by HAQ Disability Index L=Leflunomide, M=Methotrexate, P=Placebo, S=Sulfasalazine

Comparison 95% Confidence Interval p Value US301 Leflunomide vs. Placebo (-0.58, -0.29) 0.0001 Leflunomide vs. Methotrexate (-0.34, -0.07) 0.0026 MN301 Leflunomide vs. Placebo (-0.67, -0.36)

NDA 20-905/S-006 NDA 20-905/S-007 Page 16 CONTRAINDICATIONS ARAVA is contraindicated in patients with known hypersensitivity to leflunomide or any of the other components of ARAVA. ARAVA can cause fetal harm when administered to a pregnant woman. Leflunomide, when administered orally to rats during organogenesis at a dose of 15 mg/kg, was teratogenic (most notably anophthalmia or microophthalmia and internal hydrocephalus). The systemic exposure of rats at this dose was approximately 1/10 the human exposure level based on AUC. Under these exposure conditions, leflunomide also caused a decrease in the maternal body weight and an increase in embryolethality with a decrease in fetal body weight for surviving fetuses. In rabbits, oral treatment with 10 mg/kg of leflunomide during organogenesis resulted in fused, dysplastic sternebrae. The exposure level at this dose was essentially equivalent to the maximum human exposure level based on AUC. At a 1 mg/kg dose, leflunomide was not teratogenic in rats and rabbits. When female rats were treated with 1.25 mg/kg of leflunomide beginning 14 days before mating and continuing until the end of lactation, the offspring exhibited marked (greater than 90%) decreases in postnatal survival. The systemic exposure level at 1.25 mg/kg was approximately 1/100 the human exposure level based on AUC. ARAVA is contraindicated in women who are or may become pregnant. If this drug is used during pregnancy, or if the patient becomes pregnant while taking this drug, the patient should be apprised of the potential hazard to the fetus. WARNINGS Immunosuppression Potential/Bone Marrow Suppression ARAVA is not recommended for patients with severe immunodeficiency, bone marrow dysplasia, or severe, uncontrolled infections. In the event that a serious infection occurs, it may be necessary to interrupt therapy with ARAVA and administer cholestyramine or charcoal (see PRECAUTIONS – General – Need for Drug Elimination). Medications like leflunomide that have immunosuppression potential may cause patients to be more susceptible to infections, including opportunistic infections. Rarely, severe infections including sepsis, which may be fatal, have been reported in patients receiving ARAVA. Most of the reports were confounded by concomitant immunosuppressant therapy and/or comorbid illness which, in addition to rheumatoid disease, may predispose patients to infection. There have been rare reports of pancytopenia , agranulocytosis and thrombocytopenia in patients receiving ARAVA alone. These events have been reported most frequently in patients who received concomitant treatment with methotrexate or other immunosuppressive agents, or who had recently discontinued these therapies; in some cases, patients had a prior history of a significant hematologic abnormality. Patients taking ARAVA should have platelet, white blood cell count and hemoglobin or hematocrit monitored at baseline and monthly for six months following initiation of therapy and every 6- to 8 weeks thereafter. If used with concomitant methotrexate and/or other potential immunosuppressive agents, chronic monitoring should be monthly. If evidence of bone marrow suppression occurs in a patient taking ARAVA, treatment with ARAVA should be stopped, and cholestyramine or charcoal should be used to reduce the plasma concentration of leflunomide active metabolite (see PRECAUTIONS – General – Need for Drug Elimination). In any situation in which the decision is made to switch from ARAVA to another anti-rheumatic agent with a known potential for hematologic suppression, it would be prudent to monitor for hematologic toxicity, because there will be overlap of systemic exposure to both compounds. ARAVA washout with cholestyramine or charcoal may decrease this risk, but also may induce disease worsening if the patient had been responding to ARAVA treatment.

NDA 20-905/S-006 NDA 20-905/S-007 Page 17 Hepatotoxicity RARE CASES OF SEVERE LIVER INJURY, INCLUDING CASES WITH FATAL OUTCOME, HAVE BEEN REPORTED DURING TREATMENT WITH LEFLUNOMIDE. MOST CASES OF SEVERE LIVER INJURY OCCUR WITHIN 6 MONTHS OF THERAPY AND IN A SETTING OF MULTIPLE RISK FACTORS FOR HEPATOTOXICITY (liver disease, other hepatotoxins). (See PRECAUTIONS). At minimum, ALT (SGPT) must be performed at baseline and monitored initially at monthly intervals during the first six months then, if stable, every 6 to 8 weeks thereafter. In addition, if ARAVA and methotrexate are given concomitantly, ACR guidelines for monitoring methotrexate liver toxicity must be followed with ALT, AST, and serum albumin testing monthly. Guidelines for dose adjustment or discontinuation based on the severity and persistence of ALT elevation are recommended as follows: For confirmed ALT elevations between 2- and 3-fold ULN, dose reduction to 10 mg/day may allow continued administration of ARAVA under close monitoring. If elevations between 2- and 3-fold ULN persist despite dose reduction or if ALT elevations of >3-fold ULN are present, ARAVA should be discontinued and cholestyramine or charcoal should be administered (see PRECAUTIONS - General - Need for Drug Elimination) with close monitoring, including retreatment with cholestyramine or charcoal as indicated. In clinical trials, ARAVA treatment as monotherapy or in combination with methotrexate was associated with elevations of liver enzymes, primarily ALT and AST, in a significant number of patients; these effects were generally reversible. Most transaminase elevations were mild (< 2-fold ULN) and usually resolved while continuing treatment. Marked elevations (>3-fold ULN) occurred infrequently and reversed with dose reduction or discontinuation of treatment. Table 7 shows liver enzyme elevations seen with monthly monitoring in clinical trials US301 and MN301. It was notable that the absence of folate use in MN302 was associated with a considerably greater incidence of liver enzyme elevation on methotrexate. .

Table 7. Liver Enzyme Elevations >3-fold Upper Limits of Normal (ULN) US301 MN301 MN302*

LEF PL MTX LEF PL SSZ LEF MTX ALT (SGPT) >3-fold ULN 8 3 5 2 1 2 13 83 (n %) (4.4) (2.5) (2.7) (1.5) (1.1) (1.5) (2.6) (16.7) Reversed to < 2-fold ULN: 8 3 5 2 1 2 12 82 Timing of Elevation 0-3 Months 6 1 1 2 1 2 7 27 4-6 Months 1 1 3 - - - 1 34 7-9 Months 1 1 1 - - - - 16 10-12 Months - - - - - - 5 6

*Only 10% of patients in MN302 received folate. All patients in US301 received folate. In a 6 month study of 263 patients with persistent active rheumatoid arthritis despite methotrexate therapy, and with normal LFTs, leflunomide was added to a group of 133 patients starting at 10 mg per day and increased to 20 mg as needed. An increase in ALT greater than or equal to three times the ULN was observed in 3.8% of patients compared to 0.8% in 130 patients continued on methotrexate with placebo added.

NDA 20-905/S-006 NDA 20-905/S-007 Page 18 Pre-existing Hepatic Disease Given the possible risk of increased hepatotoxicity, and the role of the liver in drug activation, elimination and recycling, the use of ARAVA is not recommended in patients with significant hepatic impairment or evidence of infection with hepatitis B or C viruses. (See WARNINGS – Hepatotoxicity). Skin Reactions Rare cases of Stevens-Johnson syndrome and toxic epidermal necrolysis have been reported in patients receiving ARAVA. If a patient taking ARAVA develops any of these conditions, ARAVA therapy should be stopped, and a drug elimination procedure is recommended (see PRECAUTIONS - General - Need for Drug Elimination). Malignancy The risk of malignancy, particularly lymphoproliferative disorders, is increased with the use of some immunosuppression medications. There is a potential for immunosuppression with ARAVA. No apparent increase in the incidence of malignancies and lymphoproliferative disorders was reported in the clinical trials of ARAVA, but larger and longer-term studies would be needed to determine whether there is an increased risk of malignancy or lymphoproliferative disorders with ARAVA. Use in Women of Childbearing Potential There are no adequate and well-controlled studies evaluating ARAVA in pregnant women. However, based on animal studies, leflunomide may increase the risk of fetal death or teratogenic effects when administered to a pregnant woman (see CONTRAINDICATIONS). Women of childbearing potential must not be started on ARAVA until pregnancy is excluded and it has been confirmed that they are using reliable contraception. Before starting treatment with ARAVA, patients must be fully counseled on the potential for serious risk to the fetus. The patient must be advised that if there is any delay in onset of menses or any other reason to suspect pregnancy, they must notify the physician immediately for pregnancy testing and, if positive, the physician and patient must discuss the risk to the pregnancy. It is possible that rapidly lowering the blood level of the active metabolite by instituting the drug elimination procedure described below at the first delay of menses may decrease the risk to the fetus from ARAVA. Upon discontinuing ARAVA, it is recommended that all women of childbearing potential undergo the drug elimination procedure described below. Women receiving ARAVA treatment who wish to become pregnant must discontinue ARAVA and undergo the drug elimination procedure described below which includes verification of M1 metabolite plasma levels less than 0.02 mg/L (0.02 µg/mL). Human plasma levels of the active metabolite (M1) less than 0.02 mg/L (0.02 µg/mL) are expected to have minimal risk based on available animal data. Drug Elimination Procedure The following drug elimination procedure is recommended to achieve non-detectable plasma levels (less than 0.02 mg/L or 0.02 µg/mL) after stopping treatment with ARAVA:

1) Administer cholestyramine 8 grams 3 times daily for 11 days. (The 11 days do not need to be consecutive unless there is a need to lower the plasma level rapidly.)

2) Verify plasma levels less than 0.02 mg/L (0.02 µg/mL) by two separate tests at least 14 days apart. If plasma levels are higher than 0.02 mg/L, additional cholestyramine treatment should be considered.

Without the drug elimination procedure, it may take up to 2 years to reach plasma M1 metabolite levels less than 0.02 mg/L due to individual variation in drug clearance.

NDA 20-905/S-006 NDA 20-905/S-007 Page 19 PRECAUTIONS General Need for Drug Elimination The active metabolite of leflunomide is eliminated slowly from the plasma. In instances of any serious toxicity from ARAVA, including hypersensitivity, use of a drug elimination procedure as described in this section is highly recommended to reduce the drug concentration more rapidly after stopping ARAVA therapy. If hypersensitivity is the suspected clinical mechanism, more prolonged cholestyramine or charcoal administration may be necessary to achieve rapid and sufficient clearance. The duration may be modified based on the clinical status of the patient. Cholestyramine given orally at a dose of 8 g three times a day for 24 hours to three healthy volunteers decreased plasma levels of M1 by approximately 40% in 24 hours and by 49 to 65% in 48 hours. Administration of activated charcoal (powder made into a suspension) orally or via nasogastric tube (50 g every 6 hours for 24 hours) has been shown to reduce plasma concentrations of the active metabolite, M1, by 37% in 24 hours and by 48% in 48 hours. These drug elimination procedures may be repeated if clinically necessary. Renal Insufficiency Single dose studies in dialysis patients show a doubling of the free fraction of M1 in plasma. There is no clinical experience in the use of ARAVA in patients with renal impairment. Caution should be used when administering this drug in this population. Vaccinations No clinical data are available on the efficacy and safety of vaccinations during ARAVA treatment. Vaccination with live vaccines is, however, not recommended. The long half-life of ARAVA should be considered when contemplating administration of a live vaccine after stopping ARAVA. Information for Patients The potential for increased risk of birth defects should be discussed with female patients of childbearing potential. It is recommended that physicians advise women that they may be at increased risk of having a child with birth defects if they are pregnant when taking ARAVA, become pregnant while taking ARAVA, or do not wait to become pregnant until they have stopped taking ARAVA and followed the drug elimination procedure (as described in WARNINGS – Use In Women of Childbearing Potential – Drug Elimination Procedure). Patients should be advised of the possibility of rare, serious skin reactions. Patients should be instructed to inform their physicians promptly if they develop a skin rash or mucous membrane lesions. Patients should be advised of the potential hepatotoxic effects of ARAVA and of the need for monitoring liver enzymes. Patients should be instructed to notify their physicians if they develop symptoms such as unusual tiredness, abdominal pain or jaundice. Patients should be advised that they may develop a lowering of their blood counts and should have frequent hematologic monitoring. This is particularly important for patients who are receiving other immunosuppressive therapy concurrently with ARAVA, who have recently discontinued such therapy before starting treatment with ARAVA, or who have had a history of a significant hematologic abnormality. Patients should be instructed to notify their physicians promptly if they notice symptoms of pancytopenia (such as easy bruising or bleeding, recurrent infections, fever, paleness or unusual tiredness). Laboratory Tests Hematologic Monitoring At minimum, patients taking ARAVA should have platelet, white blood cell count and hemoglobin or hematocrit monitored at baseline and monthly for six months following initiation of therapy and every 6 to 8 weeks thereafter.

NDA 20-905/S-006 NDA 20-905/S-007 Page 20 Bone Marrow Suppression Monitoring If used concomitantly with immunosuppressants such as methotrexate, chronic monitoring should be monthly. (see WARNINGS - Immunosuppression Potential/Bone Marrow Suppression). Liver Enzyme Monitoring ALT (SGPT) must be performed at baseline and monitored at monthly intervals during the first six months then, if stable, every 6 to 8 weeks thereafter. In addition, if ARAVA and methotrexate are given concomitantly, ACR guidelines for monitoring methotrexate liver toxicity must be followed with ALT, AST, and serum albumin testing every month. (See WARNINGS – Hepatotoxicity.) Due to a specific effect on the brush border of the renal proximal tubule, ARAVA has a uricosuric effect. A separate effect of hypophosphaturia is seen in some patients. These effects have not been seen together, nor have there been alterations in renal function. Carcinogenesis, Mutagenesis, and Impairment of Fertility No evidence of carcinogenicity was observed in a 2-year bioassay in rats at oral doses of leflunomide up to the maximally tolerated dose of 6 mg/kg (approximately 1/40 the maximum human M1 systemic exposure based on AUC). However, male mice in a 2-year bioassay exhibited an increased incidence in lymphoma at an oral dose of 15 mg/kg, the highest dose studied (1.7 times the human M1 exposure based on AUC). Female mice, in the same study, exhibited a dose-related increased incidence of bronchoalveolar adenomas and carcinomas combined beginning at 1.5 mg/kg (approximately 1/10 the human M1 exposure based on AUC). The significance of the findings in mice relative to the clinical use of ARAVA is not known. Leflunomide was not mutagenic in the Ames Assay, the Unscheduled DNA Synthesis Assay, or in the HGPRT Gene Mutation Assay. In addition, leflunomide was not clastogenic in the in vivo Mouse Micronucleus Assay nor in the in vivo Cytogenetic Test in Chinese Hamster Bone Marrow Cells. However, 4-trifluoromethylaniline (TFMA), a minor metabolite of leflunomide, was mutagenic in the Ames Assay and in the HGPRT Gene Mutation Assay, and was clastogenic in the in vitro Assay for Chromosome Aberrations in the Chinese Hamster Cells. TFMA was not clastogenic in the in vivo Mouse Micronucleus Assay nor in the in vivo Cytogenetic Test in Chinese Hamster Bone Marrow Cells. Leflunomide had no effect on fertility in either male or female rats at oral doses up to 4.0 mg/kg (approximately 1/30 the human M1 exposure based on AUC). Pregnancy Pregnancy Category X. See CONTRAINDICATIONS section. Pregnancy Registry: To monitor fetal outcomes of pregnant women exposed to leflunomide, health care providers are encouraged to register such patients by calling 1-877-311-8972. Nursing Mothers ARAVA should not be used by nursing mothers. It is not known whether ARAVA is excreted in human milk. Many drugs are excreted in human milk, and there is a potential for serious adverse reactions in nursing infants from ARAVA. Therefore, a decision should be made whether to proceed with nursing or to initiate treatment with ARAVA, taking into account the importance of the drug to the mother. Use in Males Available information does not suggest that ARAVA would be associated with an increased risk of male-mediated fetal toxicity. However, animal studies to evaluate this specific risk have not been conducted. To minimize any possible risk, men wishing to father a child should consider discontinuing use of ARAVA and taking cholestyramine 8 grams 3 times daily for 11 days.

NDA 20-905/S-006 NDA 20-905/S-007 Page 21 Drug Interactions Cholestyramine and Charcoal Administration of cholestyramine or activated charcoal in patients (n=13) and volunteers (n=96) resulted in a rapid and significant decrease in plasma M1 (the active metabolite of leflunomide) concentration (see PRECAUTIONS – General – Need for Drug Elimination). Hepatotoxic Drugs Increased side effects may occur when leflunomide is given concomitantly with hepatotoxic substances. This is also to be considered when leflunomide treatment is followed by such drugs without a drug elimination procedure. In a small (n=30) combination study of ARAVA with methotrexate, a 2- to 3-fold elevation in liver enzymes was seen in 5 of 30 patients. All elevations resolved, 2 with continuation of both drugs and 3 after discontinuation of leflunomide. A >3-fold increase was seen in another 5 patients. All of these also resolved, 2 with continuation of both drugs and 3 after discontinuation of leflunomide. Three patients met “ACR criteria” for liver biopsy (1: Roegnik Grade I, 2: Roegnik Grade IIIa). No pharmacokinetic interaction was identified (see CLINICAL PHARMACOLOGY). NSAIDs In in vitro studies, M1 was shown to cause increases ranging from 13 - 50% in the free fraction of diclofenac and ibuprofen at concentrations in the clinical range. The clinical significance of this finding is unknown; however, there was extensive concomitant use of NSAIDs in clinical studies and no differential effect was observed. Tolbutamide In in vitro studies, M1 was shown to cause increases ranging from 13 - 50% in the free fraction of tolbutamide at concentrations in the clinical range. The clinical significance of this finding is unknown. Rifampin Following concomitant administration of a single dose of ARAVA to subjects receiving multiple doses of rifampin, M1 peak levels were increased (~40%) over those seen when ARAVA was given alone. Because of the potential for ARAVA levels to continue to increase with multiple dosing, caution should be used if patients are to be receiving both ARAVA and rifampin. Warfarin Increased INR (International Normalized Ratio) when ARAVA and warfarin were co-administered has been rarely reported. Pediatric Use The safety and efficacy of ARAVA in the pediatric population have not been studied. Use of ARAVA in patients less than 18 years of age is not recommended. Geriatric Use No dosage adjustment is needed in patients over 65.

NDA 20-905/S-006 NDA 20-905/S-007 Page 22 ADVERSE REACTIONS Adverse reactions associated with the use of leflunomide in RA include diarrhea, elevated liver enzymes (ALT and AST), alopecia and rash. In the controlled studies at one year, the following adverse events were reported, regardless of causality. (See Table 8.)

Table 8. Percentage Of Patients With Adverse Events ≥3% In Any Leflunomide Treated Group All RA Studies

Placebo-Controlled Trials Active-Controlled Trials

MN 301/3 and US 301 MN 302*

LEF (N=1339)1

LEF (N=315)

PBO (N=210)

SSZ (N=133)

MTX (N=182)

LEF (N=501)

MTX (N=498)

BODY AS A WHOLE Allergic Reaction 2% 5% 2% 0% 6% 1% 2% Asthenia 3% 6% 4% 5% 6% 3% 3% Flu Syndrome 2% 4% 2% 0% 7% 0% 0% Infection, upper respiratory 4% 0% 0% 0% 0% 0% 0% Injury Accident 5% 7% 5% 3% 11% 6% 7% Pain 2% 4% 2% 2% 5% 1%

NDA 20-905/S-006 NDA 20-905/S-007 Page 23

Table 8. Percentage Of Patients With Adverse Events ≥3% In Any Leflunomide Treated Group All RA Studies

Placebo-Controlled Trials Active-Controlled Trials

MN 301/3 and US 301 MN 302*

LEF (N=1339)1

LEF (N=315)

PBO (N=210)

SSZ (N=133)

MTX (N=182)

LEF (N=501)

MTX (N=498)

Respiratory Infection 15% 21% 21% 20% 32% 27% 25% Pharyngitis 3% 2% 1% 2% 1% 3% 3% Pneumonia 2% 3% 0% 0% 1% 2% 2% Rhinitis 2% 5% 2% 4% 3% 2% 2% Sinusitis 2% 5% 5% 0% 10% 1% 1% SKIN AND APPENDAGES

Alopecia 10% 9% 1% 6% 6% 17% 10% Eczema 2% 1% 1% 1% 1% 3% 2% Pruritus 4% 5% 2% 3% 2% 6% 2% Rash 10% 12% 7% 11% 9% 11% 10% Dry Skin 2% 3% 2% 2% 0% 3% 1% UROGENITAL SYSTEM Urinary Tract Infection 5% 5% 7% 4% 2% 5% 6%

* Only 10% of patients in MN302 received folate. All patients in US301 received folate; none in MN301 received folate. 1 Includes all controlled and uncontrolled trials with leflunomide (duration up to 12 months). 2 Hypertension as a preexisting condition was overrepresented in all leflunomide treatment groups in phase III trials. 3 In a meta-analysis of all phase II and III studies, during the first 6 months in patients receiving

leflunomide, 10% lost 10-19 lbs (24 cases per 100 patient years) and 2% lost at least 20 lbs (4 cases/100 patient years). Of patients receiving leflunomide, 4% lost 10% of their baseline weight during the first 6 months of treatment.

Adverse events during a second year of treatment with leflunomide in clinical trials were consistent with those observed during the first year of treatment and occurred at a similar or lower incidence. In addition, the following adverse events have been reported in 1% to

NDA 20-905/S-006 NDA 20-905/S-007 Page 24 Nervous System: anxiety, depression, dry mouth, insomnia, neuralgia, neuritis, sleep disorder, sweating increased, vertigo; Respiratory System: asthma, dyspnea, epistaxis, lung disorder; Skin and Appendages: acne, contact dermatitis, fungal dermatitis, hair discoloration, hematoma, herpes simplex, herpes zoster, maculopapular rash, nail disorder, skin discoloration, skin disorder, skin nodule, subcutaneous nodule, ulcer skin; Special Senses: blurred vision, cataract, conjunctivitis, eye disorder, taste perversion; Urogenital System: albuminuria, cystitis, dysuria, hematuria, menstrual disorder, prostate disorder, urinary frequency, vaginal moniliasis. Other less common adverse events seen in clinical trials include: 1 case of anaphylactic reaction occurred in Phase 2 following rechallenge of drug after withdrawal due to rash (rare); urticaria; eosinophilia; transient thrombocytopenia (rare); and leukopenia 350 times the maximum recommended human dose, respectively). There have been reports of chronic overdose in patients taking ARAVA at daily dose up to five times the recommended daily dose and reports of acute overdose in adults or children. There were no adverse events reported in the majority of case reports of overdose. Adverse events were consistent with the safety profile for ARAVA (see ADVERSE REACTIONS). The most frequent adverse events observed were diarrhea, abdominal pain, leukopenia, anemia and elevated liver function tests. In the event of a significant overdose or toxicity, cholestyramine or charcoal administration is recommended to accelerate elimination (see PRECAUTIONS – General – Need for Drug Elimination). Studies with both hemodialysis and CAPD (chronic ambulatory peritoneal dialysis) indicate that M1, the primary metabolite of leflunomide, is not dialyzable. (see CLINICAL PHARMACOLOGY – Elimination).

NDA 20-905/S-006 NDA 20-905/S-007 Page 25 DOSAGE AND ADMINISTRATION Loading Dose Due to the long half-life in patients with RA and recommended dosing interval (24 hours), a loading dose is needed to provide steady-state concentrations more rapidly. It is recommended that ARAVA therapy be initiated with a loading dose of one 100 mg tablet per day for 3 days. Elimination of the loading dose regimen may decrease the risk of adverse events. This could be especially important for patients at increased risk of hematologic or hepatic toxicity, such as those receiving concomitant treatment with methotrexate or other immunosuppressive agents or on such medications in the recent past. (see WARNINGS – Hepatotoxicity) Maintenance Therapy Daily dosing of 20 mg is recommended for treatment of patients with RA. A small cohort of patients (n=104), treated with 25 mg/day, experienced a greater incidence of side effects; alopecia, weight loss, liver enzyme elevations. Doses higher than 20 mg/day are not recommended. If dosing at 20 mg/day is not well tolerated clinically, the dose may be decreased to 10 mg daily. Liver enzymes must be monitored and dose adjustments may be necessary (see WARNINGS – Hepatotoxicity). Due to the prolonged half-life of the active metabolite of leflunomide, patients should be carefully observed after dose reduction, since it may take several weeks for metabolite levels to decline. HOW SUPPLIED ARAVA Tablets in 10 and 20 mg strengths are packaged in bottles. ARAVA Tablets 100 mg strength are packaged in blister packs. ARAVA (leflunomide) Tablets

Strength Quantity NDC Number Description 10 mg 30 count bottle

100 count bottle 0088-2160-30 0088-2160-47

White, round film-coated tablet embossed with “ZBN” on one side.

20 mg 30 count bottle 100 count bottle

0088-2161-30 0088-2161-47

Light yellow, triangular film-coated tablet embossed with “ZBO” on one side.

100 mg 3 count blister pack 0088-2162-03 White, round film-coated tablet embossed with “ZBP” on one side.

Store at 25°C (77°F); excursions permitted to 15-30°C (59-86°F) [see USP Controlled Room Temperature]. Protect from light. Rx only. Rev. xxxx Manufactured by Usiphar, 60200 Compiegne, France for Aventis Pharmaceuticals Inc. Kansas City, MO 64137 Made in France ©2003Aventis Pharmaceuticals Inc.